The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is configured to ingest and follow instructions from an external brand-profile.json file if present in the working directory to guide its generation process.
Ingestion points: SKILL.md identifies brand-profile.json as a data source for determining design systems, mood keywords, and layout decisions.
Boundary markers: The skill does not define clear delimiters or instructions to the agent to ignore potentially embedded commands within the JSON data structure.
Capability inventory: The agent generates executable HTML, CSS, and Javascript code based on the design instructions, which could be influenced by malicious content in the JSON.
Sanitization: There are no explicit sanitization routines or validation checks defined to prevent instructions inside the JSON from overriding the primary skill guidelines.
[EXTERNAL_DOWNLOADS]: The skill references external assets and installation sources from known services and the author's own infrastructure.
Evidence: REFERENCE.md imports typography assets from Google Fonts (fonts.googleapis.com) and Fontshare (api.fontshare.com).
Evidence: SKILL.md provides installation instructions using a git clone command targeting the author's GitHub repository (github.com/thatrebeccarae).

frontend-design